Although tremendous advances have been made in elucidating the genomic abnormalities that cause malignant cancer cells, currently available chemotherapy remains unsatisfactory, and the prognosis for the majority of patients diagnosed with cancer remains dismal. Most chemotherapeutic agents act on a specific molecular target thought to be involved in the development of the malignant phenotype. However, a complex network of signaling pathways regulate cell proliferation, and the majority of malignant cancers are facilitated by multiple genetic abnormalities in these pathways. Therefore, it is unlikely that a therapeutic agent that acts on one molecular target will be fully effective in curing a patient who has cancer.
Heat shock proteins (HSPs) are a class of chaperone proteins that are up-regulated in response to elevated temperature and other environmental stresses, such as ultraviolet light, nutrient deprivation, and oxygen deprivation. HSPs act as chaperones to other cellular proteins (called client proteins) and facilitate their proper folding and repair, and aid in the refolding of misfolded client proteins. There are several known families of HSPs, each having its own set of client proteins. The Hsp90 family is one of the most abundant HSP families, accounting for about 1-2% of proteins in a cell that is not under stress and increasing to about 4-6% in a cell under stress. Inhibition of Hsp90 results in degradation of its client proteins via the ubiquitin proteasome pathway. Unlike other chaperone proteins, the client proteins of Hsp90 are mostly protein kinases or transcription factors involved in signal transduction, and a number of its client proteins have been shown to be involved in the progression of cancer. Examples of Hsp90 client proteins that have been implicated in the progression of cancer are described below.
Her-2 is a transmembrane tyrosine kinase cell surface growth factor receptor that is expressed in normal epithelial cells. Her2 has an extracellular domain that interacts with extracellular growth factors and an internal tyrosine kinase portion that transmits the external growth signal to the nucleus of the cell. Her2 is overexpressed in a significant proportion of malignancies, such as breast cancer, ovarian cancer, prostate cancer, and gastric cancers, and is typically associated with a poor prognosis.
Akt kinase is a serine/threonine kinase which is a downstream effector molecule of phosphoinositide 3-kinase and is involved in protecting the cell from apoptosis. Akt kinase is thought to be involved in the progression of cancer because it stimulates cell proliferation and suppresses apoptosis.
Cdk4/cyclin D complexes are involved in phosphorylation of retinoblastoma protein which is an essential step in progression of a cell through the G1 phase of the cell cycle. Disruption of Hsp90 activity has been shown to decrease the half life of newly synthesized Cdk4.
The Raf family of proto-oncogenes (A-raf, B-raf and C-raf) was first identified when C-raf (raf-1) was discovered due to its homology with v-raf, the transforming gene of the mouse sarcoma virus 3611. A-raf was later discovered by screening a cDNA library under low stringency conditions using a v-raf probe, and B-raf was discovered due to its homology with C-Rmil, a transforming gene in avaian retrovirus Mill Hill No. 2. The Raf family of proteins is involved in the Ras/Raf/MEK/ERK pathway, referred to herein as the “MAP kinase pathway” (MEK stands for “MAPK/ERK kinase” and ERK stands for “extracellularly regulated kinases”), which has been implicated in the genesis and progression of many human cancers through upregulation of cell division and proliferation. All raf proteins are serine/theronine kinases which are capable of activating the MAP kinase pathway. However, B-raf is far more potent at activating this pathway than A-raf or C-raf, and mutations in the gene encoding B-raf are more common in cancer. For example, B-raf mutations have been identified in 60% to 70% of malignant melanomas, 83% of anaplastic thyroid carcinoma, 35% to 69% of papillary thyroid caricinoma, 4% to 16% of colon cancer, 63% of low-grade ovarian carcinoma, 15% of Barrett's esophageal carcinoma, 4% of acute myeloid leukemia, 3-4.8% of head and neck squamous cell carcinoma, 2%-3% of non-small-cell lung cancer, 2% of gastric carcinoma, 2% of non-Hodgkins lymphoma and has been reported in glioma, saroma, breast cancer, cholangiocarcinoma, and liver cancer. Most mutations in B-raf that have been found in human cancers are point mutations that occur in the kinase domain and are clustered in exons 11 and 15 of the gene which contains several regulatory phosphorylation sites (S446, S447, D448, D449, T599, and S602). (Beeram, et al., Journal of Clinical Oncology (2005), 23(27):6771-6790). The most prevalent mutation is the T1799A transversion mutation which accounts for more than 80% of mutations in the BRAF gene and results in a V600E mutation in B-raf. The V600E was formerly designated V599E (the gene mutation was designated T1796A) due to a mistake in the GenBank nucleotide sequence NM 004333. The corrected GenBank sequence is NT 007914 and designates the protein mutation as V600E and the gene mutation as T1799A. This corrected numbering will be used herein. This mutation is thought to mimic phosphorylation in the activation segment of B-raf since it inserts a negatively charged residue near two activating phosphorylation sites, T599 and S602, and thus results in constitutively active B-raf in a Ras independent manner. (Xing, M., Endocrine-Related Cancer (2005), 12:245-262).
Treatment of cancer cells with 17AAG, an Hsp90 inhibitor, has been shown to stimulate the degradation of B-raf, and mutant forms of B-raf have been shown to be more sensitive to degradation than the wild type. For example, when melanoma cell line A375 which contain the V600E mutation was treated with 17AAG, B-raf was degraded more rapidly than in CHL cells which contained wild type B-raf. Other B-raf mutants (e.g., V600D, G469A, G469E, G596R, G466V, and G594V) were a found to be degraded more rapidly than wild type B-raf when transvected into COS cells. However, B-raf mutants E586K and L597V were not sensitive to degradation when cells were treated with 17AAG. Therefore, it is believed that wild type B-raf in its activated form is a client protein of Hsp90 and that most mutated forms of B-raf are more dependent on Hsp90 for folding, stability and/or function than the wild type protein. (Dias, et al., Cancer Res. (2005), 65(23): 10686-10691).
Raf-1 is a MAP 3-kinase (MAP3K) which when activated can phosphorylate and acitivate the serine/threonine specific protein kinases ERK1 and ERK2. Activated ERKs play an important role in the control of gene expression involved in the cell division cycle, apoptosis, cell differentiation and cell migration.
Anaplastic large-cell lymphoma (ALCL) is a type of non-Hodgkin's lymphoma characterized by the expression of CD30/Ki-1 antigen. ALCL normally arises from T-cells, however, a subset of cases have either a null cell or B-cell phenotype. Cases that arise from B-cells are sometimes categorized as diffuse large B-cell lymphomas. About 60% of the ALCL case that express CD30/Ki-1 antigen also have the chromosomal translocation t(2;5)(p23; q35) which fuses the nucleophosmin (NPM/B23) gene to the anaplastic lymphoma kinse (ALK) gene and results in the oncogenetic fusion protein NPM-ALK with tyrosine kinase activity. Within specific subtypes of ALCL, ALK rearrangements have been observed in the following percentages: 1) 30% to 50% of pleomorphic ALCL, 2) more than 80% of monomorphic ALCL, 3) 75% to 100% of small-cell cases, and 4) 60% to 100% of lymphohistiocytic ALCL. NPM-ALK is able to transform both fibroblasts, hematopoietic, and primary bone marrow cell lines, and is thought to stimulate mitosis through the RAS pathway and the through activation of phospholipase C-gamma (PLC-gamma), and to protect against apoptosis through its activation of phosphatidylinositol 3 kinase (PI-3 kinase) survival pathway. (Duyster, et al., Oncogene (2001), 20:5623-5637). NPM-ALK has been shown to associate with Hsp90 and incubation of NPM-ALK expressing ALCL cells with the benzoquinone ansamycin, 17AAG, has been shown to disrupt this association resulting in increased degradation of NPM-ALK and induce cell-cycle arrest and apoptosis. (Georgakis, et al., Exp. Hematology (2006), 34(12):1670-1679; Bonvini, et al., Cancer Research (2002), 62:1559-1566).
The transforming protein of Rous sarcoma virus, v-src, is a prototype of an oncogene family that induces cellular transformation (i.e., tumorogenesis) by non-regulated kinase activity. Hsp90 has been shown to complex with v-scr and inhibit its degradation.
Hsp90 is required to maintain steroid hormone receptors in a conformation capable of binding hormone with high affinity. Inhibition of the action of Hsp90 therefore is expected to be useful in treating hormone-associated malignancies such as breast cancer.
p53 is a tumor suppressor protein that causes cell cycle arrest and apoptosis. Mutation of the p53 gene is found in about half of all human cancers making it one of the most common genetic alterations found in cancerous cells. In addition, p53 mutation is associated with a poor prognosis. Wild-type p53 has been shown to interact with Hsp90, but mutated p53 forms a more stable association than wild-type p53 as a result of its misfolded conformations. A stronger interaction with Hsp90 protects the mutated protein form normal proteolytic degradation and prolongs its half-life. In a cell that is heterozygous for mutated and wild-type p53, inhibition of the stabilizing effect of Hsp90 causes mutant p53 to be degraded and restores the normal transcriptional activity of wild-type p53.
Hif-1α is a hypoxia-inducible transcription factor that is up-regulated under low oxygen conditions. Under normal oxygen conditions Hif-1a associates with Von Hippel-Lindau (VHL) tumor suppressor protein and is degraded. Low oxygen conditions inhibit this association and allows Hif-1α to accumulate and complex with Hif-1β to form an active transcription complex that associates with hypoxia-response elements to activate the transcription of vascular endothelial growth factor (VEGF). Increased Hif-1α is associated with increased metastasis and a poor prognosis.
There are two classes of PKs: protein tyrosine kinases (PTKs), which catalyze the phosphorylation of tyrosine kinase residues, and the serine-threonine kinases (STKs), which catalyze the phosphorylation of serine or threonine residues. Growth factor receptors with PTK activity are known as receptor tyrosine kinases. Receptor tyrosine kinases are a family of tightly regulated enzymes, and the aberrant activation of various members of the family is one of the hallmarks of cancer. The receptor tyrosine kinase family can be divided into subgroups that have similar structural organization and sequence similarity within the kinase domain.
Epidermal Growth Factor Receptor (EGFR) is a member of the type 1 subgroup of receptor tyrosine kinase family of growth factor receptors, which play critical roles in cellular growth, differentiation, and survival. Activation of these receptors typically occurs via specific ligand binding which results in hetero- or homodimerization between receptor family members, with subsequent autophosphorylation of the tyrosine kinase domain. Specific ligands which bind to EGFR include epidermal growth factor (EGF), transforming growth factor α (TGFα, amphiregulin and some viral growth factors. Activation of EGFR triggers a cascade of intracellular signaling pathways involved in both cellular proliferation (the ras/raf/MAP kinase pathway) and survival (the PI3 kinase/Akt pathway). Members of this family, including EGFR and HER2, have been directly implicated in cellular transformation.
A number of human malignancies are associated with aberrant or overexpression of EGFR and/or overexpression of its specific ligands (Gullick, Br. Med. Bull. (1991), 47:87-98; Modijtahedi and Dean, Int. J. Oncol. (1994), 4:277-96; Salomon, et al., Crit. Rev. Oncol. Hematol. (1995); 19:183-232, the entire teachings of each of these references are incorporated herein by reference). Aberrant or overexpression of EGFR has been associated with an adverse prognosis in a number of human cancers, including head and neck, breast, colon, prostate, lung (e.g., NSCLC, adenocarcinoma and squamous lung cancer), ovaries, gastrointestinal cancers (gastric, colon, pancreatic), renal cell cancer, bladder cancer, glioma, gynecological carcinomas, and prostate cancer. In some instances, overexpression of tumor EGFR has been correlated with both chemoresistance and a poor prognosis (Lei, et al., Anticancer Res. (1999), 19:221-8; Veale, et al., Br. J. Cancer (1993); 68:162-5, the entire teachings of each of these references are incorporated herein by reference).
Gefitinib, a chemotherapeutic agent that inhibits the activity of EGFR, has been found to be highly efficacious in a subset of lung cancer patients that have mutations in the tyrosine kinase domain of EGFR. In the presence of EGF, these mutants displayed two to three times higher activity than wild type EGFR. In addition, wild type EGFR was internalized by the cells and down-regulated after 15 minutes, where as mutant EGFR was internalized more slowly and continued to be activated for up to three hours (Lynch, et al., The New England Journal of Medicine (2006), 350:2129-2139, the entire teachings of which are incorporated herein by reference).
Gliomas are another type of cancer that is characterized by amplification and/or mutation of the EGFR gene. One of the most common mutations in the EGFR gene is a deletion of exons 2-7 which results in a truncated form of EGFR in which amino acids 6-273 of the extracellular domain are replaced with a single glycine residue. This mutation is called EGFRvIII and is expressed in about half of all glioblastomas. EGFRvIII is unable to bind EGF and TGF and has constitutive, ligand-independent tyrosine kinase activity. Hsp90 co-purifies with EGFRvIII indicating that Hsp90 complexes with EGFRvIII. Moreover, Hsp90 inhibitor geldanamycin, a benzoquinone ansamycin antibiotic, was able to decrease the expression of EGFRvIII indicating that interaction with Hsp90 is essential to maintain high expression levels of EGFRvIII (Lavictoire, et al., Journal of Biological Chemistry (2003), 278(7):5292-5299, the entire teachings of which are incorporated herein by reference). These results demonstrate that inhibiting the activity of Hsp90 is an effective strategy for treating cancers that are associated with inappropriate EGFR activity.
The members of the type III group of receptor tyrosine kinases include platelet-derived growth factor (PDGF) receptors (PDGF receptors alpha and beta), colony-stimulating factor (CSF-1) receptor (CSF-1 R, c-Fms), Fms-like tyrosine kinase (FLT3), and stem cell factor receptor (c-kit). FILT3 is primarily expressed on immature hematopoietic progenitors and regulates their proliferation and survival.
Hematologic cancers, also known as hematologic or hematopoietic malignancies, are cancers of the blood or bone marrow; including leukemia and lymphoma. Acute myelogenous leukemia (AML) is a clonal hematopoietic stem cell leukemia that represents about 90% of all acute leukemias in adults with an incidence of 3.9 per 100,000 (See e.g., Lowenberg et al., N. Eng. J. Med. 341: 1051-62 (1999) and Lopesde Menezes, et al, Clin. Cancer Res. (2005), 11(14):5281-5291, the enter teachings of both references are incorporated by reference). While chemotherapy can result in complete remissions, the long term disease-free survival rate for AML is about 14% with about 7,400 deaths from AML each year in the United States. Approximately 70% of AML blasts express wild type FLT3 and about 25% to about 35% express FLT3 kinase receptor mutations which result in constitutively active FLT3. Two types of activating mutations have been identified in AML patients: internal tandem duplications (ITDs) and point mutation in the activating loop of the kinase domain. FLT3-ITD mutations in AML patients is indicative of a poor prognosis for survival, and in patients who are in remission, FLT3-ITD mutations are the most significant factor adversely affecting relapse rate with 64% of patients having the mutation relapsing within 5 years (see Current Pharmaceutical Design (2005), 11:3449-3457, the entire teachings of which are incorporated herein by reference). The prognostic significance of FLT3 mutations in clinical studies suggests that FLT3 plays a driving role in AML and may be necessary for the development and maintenance of the disease.
Mixed Lineage Leukemia (MLL) involve translocations of chromosome 11 band q23 (11 q23) and occur in approximately 80% of infant hematological malignancies and 10% of adult acute leukemias. Although certain 11q23 translocation have been shown to be essential to immortalization of hematopoietic progenitors in vitro, a secondary genotoxic event is required to develop leukemia. There is a strong concordance between FLT3 and MLL fusion gene expression, and the most consistently overexpressed gene in MLL is FLT3. Moreover, it has been shown that activated FLT3 together with MLL fusion gene expression induces acute leukemia with a short latency period (see Ono, et al., J. of Clinical Investigation (2005), 115:919-929, the entire teachings of which are incorporated by reference). Therefore, it is believed that FLT3 signally is involved in the development and maintenance of MLL (see Armstrong, et al., Cancer Cell (2003), 3:173-183, the entire teachings of which are incorporated herein by reference).
The FLT3-ITD mutation is also present in about 3% of cases of adult myelodysplastic syndrome and some cases of acute lymphocytic leukemia (ALL) (Current Pharmaceutical Design (2005), 11:3449-3457).
FLT3 has been shown to be a client protein of Hsp90, and 17AAG, a benzoquinone ansamycin antibiotic that inhibits Hsp90 activity, has been shown to disrupts the association of Flt3 with Hsp90. The growth of leukemia cell that express either wild type FLT3 or FLT3-ITD mutations was found to be inhibited by treatment with 17″AAG (Yao, et al., Clinical Cancer Research (2003), 9:4483-4493, the entire teachings of which are incorporated herein by reference).
c-Kit is a membrane type III receptor protein tyrosine kinase which binds Stem Cell Factor (SCF) to its extraellular domain. c-Kit has tyrosine kinase activity and is required for normal hematopoiesis. However, mutations in c-kit can result in ligand-independent tyrosine kinase activity, autophosphorylation, and uncontrolled cell proliferation. Aberrant expression and/or activation of c-Kit has been implicated in a variety of pathologic states. For example, evidence for a contribution of c-Kit to neoplastic pathology includes its association with leukemias and mast cell tumors, small cell lung cancer, testicular cancer, and some cancers of the gastrointestinal tract and central nervous system. In addition, c-Kit has been implicated in playing a role in carcinogenesis of the female genital tract sarcomas of neuroectodermal origin, and Schwann cell neoplasia associated with neurofibromatosis. (Yang et al., J Clin Invest. (2003), 112:1851-1861; Viskochil, J Clin Invest. (2003), 112:1791-1793, the entire teachings of each of these reference are incorporated herein by reference). c-Kit has been shown to be a client protein of Hsp90, and Hsp90 inhibitor 17AAG, a benzoquinon ansamycin, has been shown to induce apoptosis in Kasumi-1 cells, an acute myeloid leukemia cell line that harbors a mutation in c-kit.
c-Met is a receptor tyrosine kinase that is a client protein of Hsp90 and is encoded by the Met protooncogene. Hepatocyte growth factor (HGF) (also referred to as scatter factor (SF)) is the natural ligand of c-Met which binds to c-Met and leads to a variety of cellular responses such as proliferation, survival, angiogenesis, wound healin, tissue regeneration, scattering, motility, invasion and branching morphogenesis (Ma et al., Cancer and Metastasis Reviews (2003), 22: 309-325). c-Met and HGF are expressed in numerous tissues, although their expression is normally confined predominantly to cells of epithelial and mesenchymal origin, respectively. c-Met and HGF are required for normal mammalian development and have been shown to be important in cell migration, cell proliferation and survival, morphogenic differentiation, and organization of 3-dimensional tubular structures (e.g., renal tubular cells, gland formation, etc.). However, dysregulation of c-Met and/or HGF is believed to contribute to tumor growth, dissemination and invasion in several human cancers. c-Met and/or HGF are highly expressed in numerous cancers and their expression correlates with poor prognosis (Christensen, et al., Cancer Research (2003), 63:7345-7355). For example, c-Met receptor mutations have been shown to be expressed in a number of human cancers including hereditary and sporadic human papillary renal carcinomas, ovarian cancer, childhood hepatocellular carcinoma, metastatic head and neck squamous cell carcinomas, esophageal cancer and gastric cancer. Met gene amplification and over expression of c-Met has been shown to be associated with both non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC), as well as colorectal cancer, and the Tpr/Met fusion protein has been shown to be present in human osteogenic sarcoma and gastric cancer. Families with germine mutations that activate c-Met kinase are prone to multiple kidney tumors as well as tumors in other tissues. Numerous studies have correlated the expression of c-Met and/or HGF with the state of disease progression of different types of cancer (including lung, colon, breast, prostate, liver, pancreas, brain, kidney, ovarian, stomach, skin, and bone cancers).
The validity of targeting receptor tyrosine kinases (RTK) that are dysregulated in human cancers is illustrated by the successes of Gleevec targeting Bcr-Abl in chronic myelogenous leukemia and c-Kit in gastroinstinal stromal tumors, Herceptin in Her-2 overexpressing breast cancers, and Iressa in select NSCLC that have dysregulated EGFR. Compelling evidence exists for targeting c-Met in the treatment of human cancers and several small drug molecules that inhibit c-Met are currently in development. However, therapies that target specific RTK often work well initially for treating cancer but eventually fail due to additional mutations which allow RTK to maintain its activity in the presence of the drug. Moreover, the selective c-Met inhibitor SU11274, while highly affected against wild type c-Met and some mutants of c-Met, has been shown to be ineffective against other c-Met mutants (Berthou, et al., Oncogene (2004), 23:5387-5393). Therefore, a need exists to develop new anticancer therapeutics that reduce the expression and/or activity of c-Met via a different mechanism than therapeutics that directly inhibit c-Met.
BCR-ABL is an ocoprotein with tyrosine kinase activity and has been associated with chronic myelogenous leukemia (CML), with a subset of patients with acute lymphocytic leukemia (ALL) and with a subset of patients with acute myelogenous leukemia (AML). In fact, the BCR-ABL oncogene has been found in at least 90-95% of patients with CML, 20% of adults with ALL, 5% of children with ALL, and in about 2% of adults with AML. The BCR-ABL oncoprotein is generated by the transloction of gene sequences from the c-ABL protein tyrosine kinase on chromosome 9 into the BCR sequences on chromosome 22, producing the Philadelphia chromosome. The BCR-ABL gene has been shown to produce at least three alternative chimeric proteins, p230 Bcr-Abl, p210 Bcr-Abl, and p190 Bcr-Abl which have unregulated tyrosine kinase activity. The p210 Bcr-Abl fusion protein is most often associated with CML, while the p190 Bcr-Abl fusion protein is most often associated with ALL. Bcr-Abl has also been associated with a variety of additional hematological malignancies including granulocytic hyperplasia, myelomonocytic leukemia, lymphomas and erythroid leukemia.
Studies have shown that lowering the expression or activity of Bcr-Abl is effective in treating Bcr-Abl-positive leukemias. For example, agents such as As2O3 which lower Bcr-Abl expression have been shown to be highly effective against Bcr-Abl leukemias. In addition, inhibition of Bcr-Abl tyrosine kinase activity by Imatinib (also known as STI571 and Gleevic) induces differentiation and apoptosis and causes eradication of Bcr-Abl positive leukemia cells both in vivo and in vitro. In patients with CML in the chronic phase, as well as in a blast crisis, treatment with Imatinib typically will induce remission. However, in many cases, particularly in those patients who were in a blast crisis before remission, the remission is not durable because the Bcr-Abl fusion protein develops mutations that cause it to be resistence to Imatinib. (See Nimmanapalli, et al., Cancer Research (2001), 61:1799-1804; and Gorre, et al., Blood (2002), 100:3041-3044, the entire teachings of each of these references are incorporated herein by reference).
Bcr-Abl fusion proteins exist as complexes with Hsp90 and are rapidly degraded when the action of Hsp90 is inhibited. It has been shown that geldanamycin, a benzoquinone ansamycin antibiotic that disrupts the association of Bcr-Abl with Hsp90, results in proteasomal degradation of Bcr-Abl and induces apoptosis in Bcr-Abl leukemia cells.
Hsp90 has been shown by mutational analysis to be necessary for the survival of normal eukaryotic cells. However, Hsp90 is over expressed in many tumor types indicating that it may play a significant role in the survival of cancer cells and that cancer cells may be more sensitive to inhibition of Hsp90 than normal cells. For example, cancer cells typically have a large number of mutated and overexpressed oncoproteins that are dependent on Hsp90 for folding. In addition, because the environment of a tumor is typically hostile due to hypoxia, nutrient deprivation, acidosis, etc., tumor cells may be especially dependent on Hsp90 for survival. Moreover, inhibition of Hsp90 causes simultaneous inhibition of a number of oncoproteins, as well as hormone receptors and transcription factors making it an attractive target for an anti-cancer agent. In fact, benzoquinone ansamycins, a family of natural products that inhibit Hsp90, has shown evidence of therapeutic activity in clinical trials.
Although promising, benzoquinone ansamycins, and their derivatives, suffer from a number of limitations. For example, they have low oral bioavailability, and their limited solubility makes them difficult to formula. In addition, they are metabolized by polymorphic cytochrome P450 CYP3A4 and are a substrate for P-glycoprotein export pump involved in the development of multidrug resistance. Therefore, a need exist for new therapeutics that improve the prognosis of cancer patients and that reduces or overcomes the limitations of currently used anti-cancer agents.